Printer Friendly

Monitoring cocaine use and abstinence among cocaine users for contingency management interventions.

Cocaine use is a disorder of choice that has serious negative health and social consequences for both the individual user and society. Currently, there are no FDA-approved medications for treatment of cocaine dependence (Shorter, Domingo, & Kosten, 2015), making psychosocial approaches central to treating cocaine-using patients. Contingency management interventions are evidence-based psychosocial treatments that have been highly effective in reducing use of a wide range of substances, including cocaine (Knapp, Soares, Farrell, & Silva de Lima, 2007; Lussier, Heil, Mongeon, Badger, & Higgins, 2006; Minozzi, Saulle, De Crescenzo, & Amato, 2015). Under these types of interventions, abstinence reinforcement is arranged by routinely collecting biological measures of drug use and then delivering incentives (e.g., money or vouchers) when the measure indicates drug abstinence, and withholding incentives when the measure indicates drug use. One of the most effective applications of the procedure uses an escalating schedule of reinforcement with a reset contingency, in which the value of the incentive increases with each consecutive drug-negative urine sample. A drug-positive or missed urine sample results in loss of the incentive and a reset in the value of the next available incentive to the initial value (Higgins et al., 1991).

Detection of benzoylecgonine, the primary metabolite of cocaine, is a common method for monitoring cocaine use during contingency management interventions. Qualitative urinalysis testing (i.e., testing that solely identifies whether a particular substance is present or absent) is often used, with results expressed as positive or negative if the benzoylecgonine concentration is above or below, respectively, a certain cutoff value (typically 300 ng/ml). Urine samples are often collected two or three times per week (every 48 to 72 hours) because it has been reported that benzoylecgonine can be detected in urine for approximately 48 hours after cocaine use (for a review, see Verstraete, 2004). However, the duration of the window of detection may vary based on several factors, such as the amount of cocaine used, route of administration, the amount of water consumed, and individual differences in drug metabolism and excretion (for a review, see Cone & Dickerson, 1993). Because of this variability in the window of detection, it is possible that a single instance of cocaine use may result in multiple cocaine-positive urine samples if subsequent samples are collected before all of the metabolite has been excreted (Preston, Silverman, Schuster, & Cone, 1997; Preston, Goldberger, & Cone, 1998; Preston, Epstein, Cone, Wtsadik, Huestis, & Moolchan, 2002). This carryover effect could be problematic for contingency management interventions: in some cases, participants may stop using cocaine, yet will not receive an incentive because not enough time has elapsed before urine sample collection. Simply testing less frequently so as to reduce the likelihood of carryover effects may be problematic because it could allow new instances of cocaine use and abstinence to go undetected (for a discussion, see Cone & Dickerson, 1993).

Preston and colleagues (1997) developed procedures to identify whether a urine sample is cocaine-positive due to carryover from earlier use or due to recent cocaine use. Development of these "New Use" criteria was based on pharmacokinetic parameters of cocaine and benzoylecgonine, and used quantitative urinalysis testing (i.e., testing that identifies how much of a substance is present). According to these criteria, when the current and prior urine samples have been collected at least 48 hours apart and have benzoylecgonine concentrations that are greater than 300 ng/ml, the current sample may be considered positive due to carryover if the benzoylecgonine concentration of the current sample has decreased by 50% or more from the prior sample. These criteria for detecting new use versus carryover from previous drug use were validated using data from a laboratory-based cocaine dosing study and a clinical trial evaluating a treatment for cocaine use (Preston et al., 1997). In the latter, the New Use criteria were applied to data from a 12-week clinical trial in which cocaine-using methadone patients were randomly assigned to a control group, or a group that could earn vouchers for providing cocaine-negative urine samples (benzoylecgonine concentrations [less than or equal to] 300 ng/ml). The percentage of urine samples identified as positive due to new use of cocaine when the New Use criteria were applied was lower than the percentage of urine samples testing positive under the conventional qualitative approach. For example, in the voucher group, about 30% of urine samples were cocaine-positive when the New Use criteria were applied versus about 45% when qualitative urinalysis testing was used, suggesting that the conventional qualitative approach may have missed opportunities to reinforce cocaine abstinence.

Although arranging reinforcement for cocaine abstinence using the New Use criteria could allow for more immediate reinforcement of the initiation of cocaine abstinence and improve the effectiveness of the intervention, most contingency management interventions have used qualitative urinalysis testing. In a recent review of contingency management interventions targeting substance use disorders (Davis, Kurti, Skelly, Redner, White, & Higgins, 2016), 25 recent studies arranged reinforcement for cocaine abstinence using urinalysis testing and 23 (92%) used qualitative testing. Of those studies that reported the average percentage of urine samples that were negative for cocaine (16 of the 23 studies), 39% on average (median) were cocaine-positive. This is consistent with previous reports suggesting that contingency management interventions are effective in promoting cocaine abstinence in many but not all participants (Higgins. Silverman. & Heil, 2008). While we cannot know why the interventions did not promote cocaine abstinence in some participants, we suggest that further development of methods that may allow for the precise reinforcement of recent cocaine abstinence is warranted.

The present secondary analysis had two primary objectives. First, we sought to identify changes in benzoylecgonine concentrations that could be indicative of cocaine abstinence and thus available for reinforcement. Second, the optimal criterion to identify decreases in benzoylecgonine that are indicative of cocaine abstinence (e.g., a decrease of 50% or more from the prior sample) has not been definitively established; we sought to provide further information on the topic. This study examined data from a controlled clinical trial of an intervention that arranged reinforcement of cocaine abstinence among cocaine-using methadone patients, a population at risk for poor treatment retention, continued opioid use, and an increased probability of relapse due to their cocaine use (Leri, Bruneau, & Stewart, 2003). Half of the participants in the clinical trial were exposed to a cocaine-abstinence contingency that arranged reinforcement of recent cocaine abstinence. Cocaine use was monitored for the other half of the participants, but no consequences for cocaine use were implemented with those participants. The main findings from the original trial showed that the cocaine-abstinence contingency increased cocaine abstinence (Silverman et al., 2007). The present paper provides a more detailed examination of the changes in benzoylecgonine concentrations that occurred during the trial for participants who were exposed to the cocaine-abstinence reinforcement contingency. The overarching goal of the present paper was to aid in the development of procedures that may allow for the precise reinforcement of recent cocaine abstinence during contingency management interventions.


The original report of the clinical trial provides a full description of the methods employed in that study (Silverman et al., 2007). The present paper includes a description of the methods that arc critical to the secondary analyses being presented.

Participants and Setting

The clinical trial was conducted in a model therapeutic workplace located at the Center for Learning and Health in Baltimore, MD, USA. Participants (N = 28) were at least 18 years old, unemployed, enrolled in a methadone maintenance program, reported using cocaine in the 30 days prior to study intake, provided a cocaine-positive urine sample, reported injection drug use in the 30 days prior to study intake, and showed signs of injection drug use (i.e., track marks). Table 1 shows participant characteristics at intake.


Participants in the clinical trial were invited to attend the therapeutic workplace for a 4-week baseline period. While attending the workplace, a participant's "job" was to participate in computcr-bascd education and job-skills training. Participants could work for 4 hr every weekday and could earn $8 per hr in base pay plus about $2 per hr for their performance on the training programs. All earnings were in the form of vouchers that could be exchanged for goods and services. On Mondays, Wednesdays, and Fridays, urine samples were collected under observation and tested for cocaine; however, participants could work even if their urine samples tested cocaine-positive. At the end of the baseline period, participants who attended the workplace on at least 50% of the workdays, provided at least two cocaine-positive urine samples, and were still enrolled in methadone treatment were invited to participate in the main study period.

Main Study Period

Participants who entered the main study period could attend the workplace for an additional 26 weeks and were randomly assigned to an abstinence contingency (abstinence and work) or no abstinence contingency (work only) condition. Participants in both conditions continued to provide urine samples on Mondays, Wednesdays, and Fridays, and could earn base and performance pay. Participants in the abstinence contingency condition either had to provide urine samples with benzoylecgonine concentrations below 300 ng/ml or urine samples that indicated no recent cocaine use (i.e., their urinary benzoylecgonine concentrations had decreased by at least 20% per day since the last sample) to gain daily access to the workplace and to continue earning the maximum base pay of $8 per hr. Participants who provided a urine sample that did not meet the recent abstinence criteria or failed to provide a required sample, were not allowed to work and received a base pay reset from $8 per hr to $ 1 per hr. In addition, participants were required to provide a urine sample every workday until they provided a sample that met the abstinence requirement. After a participant's base pay was reset, it could increase by $ 1 per hr each day that the participant met the cocaine abstinence requirement and worked at least 5 min, until it reached the maximum pay rate of $8 per hr. Participants in the no abstinence contingency condition were allowed to work and earn maximum pay even if their urine samples tested cocaine-positive. In the analyses reported in this paper, we only present data on participants in the abstinence and work condition.

Urinalysis Collection and Testing

Urine samples were collected from participants under same-gender observation on Mondays, Wednesdays, and Fridays.

All samples were immediately temperature-tested following collection. A urine sample was accepted only if the temperature was within a minimum (33.3[degrees]C for women; 34.4[degrees]C for men) and maximum (37.2[degrees]C) temperature range. Urine samples were tested for the cocaine metabolite benzoylecgonine using the Abbott AxSym[R] system (Abbott Laboratories, Abbott Park, IL, USA). Benzoylecgonine concentration was quantified, and that value was compared to previous samples to determine if there was evidence of recent cocaine use. For this analysis, the sample was considered positive for cocaine if urinary benzoylecgonine concentrations exceeded 300 ng/ml.

Data Analysis

To calculate the percent by which urinary benzoylecgonine concentrations changed between two consecutive samples, the following formula was used: % change = 100-100[(curr/prev).sup.1/# days] This formula calculated the percent change per day based on the current benzoylecgonine concentration, the previously obtained benzoylecgonine concentration, and the number of days that passed between obtaining the two samples. This formula, which assumed that benzoylecgonine concentrations changed at a constant rate across the days between obtained samples, was used to implement the abstinence contingency in the main clinical trial (Silverman et al., 2007) and was based on the New Use criteria developed by Preston et al. (1997).

Periods leading to a cocaine-negative sample were identified. These included consecutively obtained samples that started with a benzoylecgonine concentration above 300 ng/ml, were then followed by one or more consecutive urine samples demonstrating decreases in benzoylecgonine concentration, with the final urine sample ending with a benzoylecgonine concentration at or below 300 ng/ml. Initial analyses were conducted to estimate the likelihood of a specific percent decrease occurring during a period leading to a cocaine-negative sample. This analysis was done by calculating the percent decrease in benzoylecgonine concentrations between consecutive urine samples (using the formula detailed above and excluding those decreases that occurred between consecutive negative samples) and separating the percent decreases into binned intervals (i.e., 0% to -9.9%, -10% to -19.9%, etc). The number of instances in each of the bins that were from periods leading to a cocaine-negative sample was then divided by the total number of values in each of the bins. Next, analyses were conducted to determine the number of days required to produce a cocaine-negative sample during these periods. A Pearson's correlation coefficient was calculated to assess the relation between the number of days required to produce a negative sample and the starting benzoylecgonine level.


Figure 1 shows changes in benzoylecgonine concentrations for consecutive urine samples in which the first sample was cocaine positive (i.e., > 300 ng/ml). The first three bars show three distinct categories of change: results are shown for consecutive urine samples in which the benzoylecgonine concentration stayed the same or increased ([greater than or equal to] 0%); decreased and transitioned from cocaine-positive to cocaine-negative (T.N.); or decreased, but was still cocaine-positive (<0%). The nine remaining bars show the percent by which the benzoylecgonine concentrations decreased; the values are presented as cumulative percentages (i.e., greater than a 10% decrease, greater than a 20% decrease, etc). Approximately 32% of the urine samples transitioned from a positive to a negative value and, therefore, would have met abstinence criteria if only qualitative urinalysis testing was used. Approximately 29% of the urine samples met the quantitative criterion for recent abstinence (i.e., at least a 20% decrease in benzoylecgonine concentration per day) and larger decreases in benzoylecgonine concentrations (at least 30%) were observed in approximately 25% of the urine samples.

As shown in Fig. 2, as the size of a sample-to-sample decrease in benzoylecgonine concentration increased, the likelihood of that sample occurring during a period leading to a cocaine-negative urine sample generally increased. About 17% of all occurrences of a 20-29.9% decrease occurred during a period leading to a cocaine-negative urine sample; however, about 53% of all occurrences of a 90% or greater decrease occurred during such a period. Thus, larger decreases in benzoylecgonine concentrations appear more indicative of eventual provision of a cocaine-negative urine sample than smaller decreases.

Figure 3 shows the number of days required to produce a cocaine-negative urine sample as a function of the starting benzoylecgonine concentration level. Only concentrations from periods leading to a cocaine-negative urine sample were included. Starting benzoylecgonine concentrations ranged from 315 ng/ml to 213,035 ng/ml. The number of days to produce a negative sample ranged from 1 to 10 days, although a majority of transitions from cocaine-positive to co-cainenegative occurred in 7 days or less. The number of days required to produce a negative sample increased as a function of, and was significantly correlated with, the starting benzoylecgonine level (r = 0.43, p < 0.001).


Cocaine use during contingency management interventions is often monitored via urinalysis detection of the primary cocaine metabolite, benzoylecgonine, and reinforcement is arranged only when a urine sample tests at or below an accepted cutoff concentration (e.g., benzoylecgonine concentration [less than or equal to] 300 ng/ml). Samples at or below the cutoff are considered cocaine-negative and are selected for reinforcement. Because cocaine may be detected in urine for several days after abstinence is initiated, however, reinforcement also can be arranged when the benzoylecgonine concentration has decreased by a pre-determined percentage from the preceding sample. The use of these qualitative versus quantitative methods for monitoring cocaine use in contingency management interventions may have important implications for the effects on drug abstinence.

The present paper sought to examine the patterns of decreases of benzoylecgonine that could be indicative of abstinence in chronic cocaine users and thus could be available for reinforcement. Changes in benzoylecgonine concentrations were examined in urine samples collected three times per week for 30 weeks from cocaine users who were exposed to a contingency in which reinforcement was delivered when urine samples suggested recent cocaine abstinence (i.e., when benzoylecgonine concentrations were [less than or equal to] 300 ng/ml or decreased by at least 20% per day since the last sample). Of the urine samples that would be considered positive by the qualitative testing criterion (i.e., benzoylecgonine concentration >300 ng/ml), approximately one third of the urine samples decreased by more than 20% per day, while approximately 25%, 21%, and 17% of the samples decreased by more than 30%, 40%, and 50% per day, respectively. These larger decreases became increasingly more likely to be associated with ultimately providing a cocaine-negative urine sample. Thus, contingencies requiring decreases larger than 20% per day could be used in future studies, although the frequency of reinforcement would likely decrease with larger requirements.

The results of the present study suggest that there is large inter-subject variability in the length of time that the cocaine metabolite benzoylecgonine can be detected in urine. Prior estimates of the window of detection of cocaine in urine have largely come from laboratory-based studies, which often administer doses of cocaine that are low compared with street doses used by many cocaine-dependent individuals (Verstraete, 2004). While it has been estimated that cocaine can be detected in urine for approximately 48 hours after cocaine use (for a review, see Verstraete, 2004), in the present study, it required an average of 88 hours (range = 24 to 240 hours) to produce a cocaine-negative urine sample following a positive sample. This result is consistent with prior research conducted with high-dose cocaine users showing that benzoylecgonine continued to be detected for much longer than 48 hours after cessation of cocaine use (Preston, Epstein, Cone, Wtsadik, Huestis, & Moolchan, 2002; Weiss & Gawin, 1988).

The main limitation of this study is that the actual amounts and patterns of cocaine use of participants are not known. For example, consecutive decreases in cocaine use leading to a cocaine-negative urine sample may not represent sustained cocaine abstinence. Nevertheless, the study does show the amounts and frequencies of different decreases in urinary benzoylecgonine concentrations in a clinical population that are available for reinforcement. Another limitation is that the present study only included methadone-maintained cocaine users; the findings may not generalize to cocaine users who are not receiving methadone treatment. Both cocaine and methadone are metabolized in part by cytochrome P450 3A4 enzymes (CYP 3A4), suggesting a potential for pharmacokinetic interaction between the two drugs. However, the influence of methadone on cocaine metabolism should be minimal because the involvement of CYP 3A4 in cocaine metabolism is relatively low (McCance-Katz, Jatlow, & Rainey, 2010).

The variability in the window of detection of cocaine in urine can have important implications when designing and implementing contingency management interventions. For example, qualitative urinalysis testing, which indicates whether a sample is cocaine-positive or cocaine-negative, is fairly simple to use and interpret, and costs less than quantitative testing. However, qualitative testing may be insensitive to moderate decreases in cocaine use due to carryover positives and may miss opportunities to reinforce abstinence initiation. Quantitative testing may be more sensitive to decreases in cocaine use; however, the selection of an appropriate criterion for distinguishing new use from carryover effects is an important consideration. If the criterion is too liberal, it is possible that decreases in cocaine use that do not reflect abstinence initiation may be reinforced. If the criterion is too stringent, opportunities that are available for reinforcement may be missed. Nevertheless, the present data provide information to aid in the assessment of what sorts of benzoylecgonine concentration decreases might reasonably be available for reinforcement in contingency management interventions.

Author Note The preparation of this publication was supported by the National Institute on Drug Abuse of the National Institutes of Health under award numbers R01DA012564, R01DA013107, R01DA019497. ROI DA037314, and T32DA07209. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Compliance with Ethical Standards

Funding This study was funded by the National Institute on Drug Abuse of the National Institutes of Health under award numbers R01DAOI2564, R0IDA013107. R01DA019497. ROI DA037314. and T32DA07209.

Conflict of Interest On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical Approval All procedures performed involving human participants were in accordance with the ethical standards of the Johns Hopkins Institutional Review Board and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed Consent Informed consent was obtained from all individual participants included in the study.


Cone. E. J., & Dickerson, S. L. (1993). Efficacy of urinalysis in monitoring heroin and cocaine abuse patterns: Implications in clinical trials for treatment of drug dependence. NIDA Research Monograph, 128, 46 46.

Davis, D. R., Kurti, A. N., Skelly, J. M., Redner, R., White, T. J.. & Higgins. S. T. (2016). A review of the literature on contingency management in the treatment of substance use disorders, 20092014. Preventive Medicine, 92, 36-46.

Higgins, S. T.. Silverman, K., & Heil, S. H. (eds.). (2008). Contingency management in substance abuse treatment. Guilford Press.

Higgins, S. T.. Delaney, D. D., Budney. A. J.. Bickel, W. K., Hughes, J. R., Foerg, F., & Fenwick, J. W. (1991). A behavioral approach to achieving initial cocaine abstinence. American Journal of Psychiatry, 148, 1218-1224.

Knapp, W. P., Soares, B., Farrell, M., & Silva de Lima, M. (2007). Psychosocial interventions for cocaine and psychostimulant amphetamines related disorders. The Cochrane Library.

Leri, F., Bruneau, J., & Stewart, J. (2003). Understanding polydrug use: Review of heroin and cocaine co-use. Addiction. 98, 7-22.

Lussier, J. P., Heil, S. H., Mongeon, J. A., Badger, G. J., & Higgins, S. T. (2006). A meta-analysis of voucher-based reinforcement therapy for substance use disorders. Addiction, 101(2), 192-203.

McCance-Katz, E. F., Jatlow, P., & Rainey, P. M. (2010). Effect of cocaine use on methadone pharmacokinetics in humans. The American Journal on Addictions, 19, 47-52.

Minozzi, S., Saulle, R., De Crescenzo, F., & Amato. L. (2015). Psychosocial interventions for psychostimulant misuse. Cochrane Database of Systematic Reviews.

Preston. K. L., Epstein. D. H., Cone. E. J., Wtsadik. A. T., Huestis, M. A., & Moolchan, E. T. (2002). Urinary elimination of cocaine metabolites in chronic cocaine users during cessation. Journal of Analytical Toxicology, 26(1), 393-400.

Preston, K. L., Goldberger. B. A., & Cone, E. J. (1998). Occurrence of cocaine in urine of substance-abuse treatment patients. Journal of Analytical Toxicology. 22(1), 580-586.

Preston, K. L.. Silverman, K., Schuster, C. R., & Cone, E. J. (1997). Assessment of cocaine use with quantitative urinalysis and estimation of new uses. Addiction, 92(6), 717-727.

Shorter, D., Domingo, C. B., & Kosten, T. R. (2015). Emerging drugs for the treatment of cocaine use disorders: A review of neurobiological targets and pharmacotherapy. Expert Opinion on Emerging Drugs, 20, 15-29.

Silverman, K., Wong, C. J., Needham, M., Diemer, K. N., Knealing, T., Crone-Todd, D., ... Kolodner. K. (2007). A randomized trial of employment-based reinforcement of cocaine abstinence in injection drug users. Journal of Applied Behavior Analysis, 40, 387-410.

Verstraete, A. G. (2004). Detection times of drugs of abuse in blood, urine, and oral fluid. Therapeutic Drug Monitoring, 26(2), 200-205.

Weiss, R. D., & Gawin, F. H. (1988). Protracted elimination of cocaine metabolites in long-term, high-dose cocaine abusers. The American Journal of Medicine, 85(6), 879-880.

DOI 10.1007/S40732-017-0236-1

[mail] August F. Holtyn

August F. Holtyn [1,2] * Todd W. Knealing [1,2,3] * Brantley P. Jarvis [1,2] * Shrindhi Subramaniam [1,2] * Kenneth Silverman [1,2]

[1] Johns Hopkins University, School of Medicine, 5200 Eastern Ave, Baltimore, MD2I224. USA

[2] Center for Learning and Health, Department of Psychiatry & Behavioral Sciences, Johns Hopkins University School of Medicine, 5200 Eastern Ave. Baltimore, MD 21224, USA

[3] Present address: Department of Psychology, Briar Cliff University, Sioux City, IA, USA

Caption: Fig. 1 The percentage of urine samples for which the benzoylecgonine (BZE) concentration did not change or increased ([greater than or equal to] 0%), decreased and transitioned from cocaine-positive to cocaine-negative (T.N.), or decreased by a specific percentage (<0% through < -90%; shown as cumulative percentages) but was still cocaine-positive. These data only include changes in benzoylecgonine concentrations for consecutive urine samples in which the first sample was cocaine-positive

Caption: Fig. 2 The percent of urine samples leading to a cocaine-negative sample (benzoylecgonine concentration [less than or equal to] 300 ng/ml) plotted as a function of their percent decrease in benzoylecgonine (BZE) level per day. The percent of urine samples in each bin was calculated by dividing the number of instances in which benzoylecgonine concentrations decreased during periods leading to a negative sample by the total number of instances in which benzoylecgonine concentrations decreased (excluding those decreases that occurred between consecutive negative samples)

Caption: Fig. 3 Number of days to produce a cocaine-negative urine sample (benzoylecgonine concentration [less than or equal to] 300 ng/ml) as a function of starting benzoylecgonine (BZE) level for urine samples that were from a period leading to a cocaine-negative urine sample
Table 1 Participant characteristics at intake (N = 28)


Age, mean (SD), years                 43.9 (6.5)

Gender, %
  Female                              64
  Male                                36

Race, %
  Black                               93
  White                               7
Married. %                            21
High school diploma or GED, %         54
Usually unemployed past 3 years, %    61
HIV-positive, %                       21
Living in poverty, %                  100
Lifetime felony conviction, %         71

DSM IV Diagnosis. %
  Heroin-dependent                    75
  Cocaine-dependent                   93
  Alcohol-dependent                   29
Days used, past 30 days, mean (SD)
  Heroin                              9.5 (10.4)
  Cocaine                             22.3 (9.2)
  Alcohol                             5.6 (9.5)
COPYRIGHT 2017 Springer
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Holtyn, August F.; Knealing, Todd W.; Jarvis, Brantley P.; Subramaniam, Shrindhi; Silverman, Kenneth
Publication:The Psychological Record
Article Type:Report
Date:Jun 1, 2017
Previous Article:A longitudinal behavioral economic analysis of non-medical prescription opioid use among college students.
Next Article:Saliva cotinine as a measure of smoking abstinence in contingency management--a feasibility study.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters